MidiPGAN: A Progressive GAN Approach to MIDI Generation
| Autor: | Kuo-Ming Chao, Guillaume Mougeot, Yan Sun, Hongming Cai, Lihong Jiang, Sebastian Walter |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
050101 languages & linguistics
MIDI Computer science business.industry Process (engineering) media_common.quotation_subject 05 social sciences Music generation 02 engineering and technology computer.file_format Musical Machine learning computer.software_genre Upsampling Metric (mathematics) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing 0501 psychology and cognitive sciences Quality (business) Encoder decoder Artificial intelligence business computer media_common |
| Zdroj: | CSCWD |
| DOI: | 10.1109/cscwd49262.2021.9437618 |
| Popis: | While recent research in music generation has mostly focused on encoder decoder architectures and self-attention mechanisms, prominent advancements regarding GANs have not yet been incorporated for the creation of music. These include solutions for major challenges when training GANs, most importantly training instability. In this work, we aim to apply this new knowledge to music generation, in order to make it more efficient and enable the automatic creation of music of higher quality. We utilize the progressive approach towards GANs, and implement it to train on symbolic music data. For best results, we process this data to obtain a new dataset, which matches the progressive approach. To achieve this, we propose a new way of downsampling fit for musical data. We furthermore conduct a user study to evaluate our results, and compute an FID score of 12.30 as objective metric. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|